Give "Call Chain" primitive an input to decide how to handle clones

[srlm]

The "Call Chain" primitive: 

Sometimes I want it to behave exactly as it does today. But sometimes, like when I'm generating error logs and trying to only write down one copy of an error, I'd like it to give me the edit-time name of VIs instead of the run-time name of VIs. That means that reentrant clones would not include the trailing number.

I would like there to be an input (bool or enum) to specify the naming of clones in the call chain. 

[wiebe@CARYA]

---

@srlm wrote:

That means that reentrant clones would not include the trailing number.

---

Not  just the trailing number. Sometimes VIs get a name as "vi name:instance:<guid>.vi".

It's not too hard to remove those with a reg.ex..

I have had the need for removing those numbers, but usually I want to be more selective. So, I don't want to remove all numbers, but only from the last 2 or 3 VIs in the chain. Usually for ignoring repeating errors.

[raphschru]

Here are some implementations of what wiebe explained:

   

  

It also removes the dot or the colon before the numbers.

[wiebe@CARYA]

That doesn't remove the :instance:<guid> from "vi name:instance:<guid>.vi".

I think this name style is typically for static clones of VIs.

[raphschru]

Do you refer to e.g. malleable subVI instance names ?

I think malleable instances are never part of the call chain or are they ?

You have an example of when that happens ?

[wiebe@CARYA]

I'd have to check. I'm not sure if these instance names can actually end up in the call chain. Could indeed be inlining (malleable) related. I see these names in heap peek. Testing this isn't that easy.